Perfluorinated compounds (PFCs) in fish from Wisconsin’s major rivers and Great Lakes Meghan C.W. Williams, Candy S. Schrank Wisconsin Department of Natural Resources 101 S Webster Street Madison WI 53707 Abstract — The Wisconsin Department of Natural Resources (WDNR) has been tracking bioaccumulating pollutants in fish that are consumed by wildlife, anglers, and anglers’ families since the 1970s. Beginning in 2006, this effort has included quantifying levels of perfluorinated compounds (PFCs) in Wisconsin fish from the Great Lakes and major river systems. The WDNR also has access to PFC data from fish collected as part of the United States Environmental Protection Agency’s 2010 National Coastal Condition Assessment/Great Lakes Human Health Fish Tissue Study. This report summarizes the concentrations of PFCs found in 28 fish species from 7 river systems and Lakes Michigan and Superior, and explores the factors affecting PFC concentrations in fish fillets. PFC contamination was found to be spatially heterogeneous, with perfluorooctane sulfonate (PFOS) present in highest concentrations and present in the highest number of samples compared to other PFCs. PFCs in fish sampled from the Great Lakes were generally lower than those sampled from riverine locations, particularly the Mississippi River, suggesting that proximity to a PFC source is an important factor affecting concentrations. Advisory concentration ranges formulated by the Minnesota Department of Health were used evaluate PFOS concentrations in Wisconsin fish. PFOS levels in most fish from most locations did not supersede Wisconsin’s general statewide advisories or advice already in place due to polychlorinated biphenyl (PCB) concentrations, although there are species from some Mississippi River locations where exceptions to general statewide advice are currently provided due to PFOS. We suggest continued monitoring of PFCs in Wisconsin fish, particularly in areas of known contamination or use. Perfluorinated compounds (PFCs) are a group biphenyls (PCBs) and polybrominated of man-made chemicals which are used for a diphenyl ethers (PBDEs), PFCs are not variety of consumer and industrial purposes. lipophilic. They instead bind to protein, In use since the 1950s (Lindstrom et al. 2011), particularly in the liver and in blood (Jones et PFCs function as stain–, oil–, and water– al. 2003; Consoer et al. 2014), meaning that PFC repellants for fabrics, carpet, cookware, and accumulation patterns and factors are paper products, and are a component of dissimilar to PCBs and PBDEs. industrial fire-fighting foams (Renner 2001; Prevedouros et al. 2006). Because of their The widespread nature of PFC contamination widespread production and use (Lau et al. is particularly troubling given the wide array 2007), and because PFCs are formulated to be of negative health effects that are associated extremely-heat stable (Bhavsar et al. 2014), with their exposure. Between 2005 and 2013, an measurable concentrations of PFCs have been epidemiological study called the C8 Health found in in nearly every corner of the world Project was conducted to determine whether (Giesy & Kannan 2001; Ahrens 2011). exposure to perfluorooctanoate (PFOA) was associated with adverse health outcomes in a In the United States, commercial formulations highly exposed population (Frisbee et al. 2009). of PFCs were produced by many companies, The C8 study found that cancers (kidney, including Arkema, Asahi, Ciba, Clariant, testicular, and thyroid), pregnancy-induced Daikin, DuPont, 3M/Dyneon, and Solvay hypertension, and ulcerative colitis were Solexis (Betts 2007). Commercial formulations linked to high levels of PFOA or contained many different PFC types having perfluorooctanesulfonate (PFOS) in the blood varying carbon chain lengths (Prevedouros et of study participants (Barry et al. 2013; Darrow al. 2006) and functional groups. PFCs’ ability to et al. 2013; Steenland et al. 2013; Watkins et al. bioaccumulate is greater if the carbon chain 2013; Stahl et al. 2014; Watkins et al. 2014). length is greater than 7, and those containing a PFOA has been additionally linked to cardiac sulfonate (perfluoroalkyl sulfonates or PFSAs) problems (Shankar et al. 2012) and cancer tend to be more bioaccumulative than those (Christensen et al. 2016) in adults. Exposure to with a carboxylate functional group (PFCAs) PFOS is tentatively associated with immune even when they have the same chain length problems (Grandjean et al. 2012) and low birth (Murphy et al. 2012). Unlike other persistent weight (Apelberg et al. 2007) in children. organic pollutants such as polychlorinated Although other adverse effects have been documented using animal models (Murphy et al. 2012), it is difficult to determine whether these effects will also occur in humans. As a result of the previously discussed evidence pointing towards PFCs’ harmful health effects and widespread global presence, manufacturers began working with the United States Environmental Protection Agency (USEPA) to phase out PFC production. 3M was the primary manufacturer of PFOS and voluntarily phased out its use in 2002 (Lindstrom et al. 2011). Other companies agreed to reduce PFOA in their products by 95% by 2010 and completely eliminate their use by 2015 (Betts 2007). Although phasing out PFC manufacture may eventually result in decreasing fish tissue concentrations (similar to what has been documented with PCBs; Rasmussen et al. 2014), fish presently represent a major PFC exposure route for humans (Jain 2014). Potential exposure due to fish consumption is a threat to Wisconsin anglers and their families, as several studies have demonstrated that PFCs can accumulate in freshwater fish to concentrations that pose a threat to human health (Martin et al. 2004; Ye et al. 2008; fish species from waterbodies near industrial Delinsky et al. 2009; Xiao et al. 2013; Stahl et al. centers and/or Great Lakes Areas of Concern 2014). (Fig. 1). This dataset includes 28 fish species collected from Lake Michigan, Lake Superior, In order to asses the threat of PFC exposure to and the Fox, Menominee, Milwaukee, those who consume fish caught in Wisconsin Mississippi (Pools 3, 4, and 5, 5A, and 6), waters, we present here PFC concentrations Peshtigo, St. Louis, and Wisconsin Rivers. measured in fillets of freshwater fish collected by the Wisconsin Department of Natural Fish samples collected by the WDNR were Resources (WDNR) and the USEPA from wrapped in aluminum foil and frozen in the Wisconsin’s major river systems and Great field before transport to the Wisconsin State Lakes between 2006 and 2012. Similar to Laboratory of Hygiene (WSLH) in Madison, previous research on PFCs in freshwater fish WI. At the WSLH, thawed whole fish or fillets (Ahrens & Bundschuh 2014), we expected to (depending on species) were homogenized and find PFOS most frequently and in highest subsamples were refrozen until processing. concentrations in fillet samples. Further, we expected that concentrations of both PFOS and Partially thawed fish homogenates were total PFCs (ΣPFCs) would be spatially quantified for up to 17 PFCs according to heterogeneous, reflecting possible local usage methods developed by Ye et al. (2008). Due to or inputs to Wisconsin waters. changes in instrument sensitivity (Table 1), different numbers of PFC types were Sampling and analysis quantified in different years. Furthermore, all This dataset includes fish samples collected by fish samples analyzed within the same both the WDNR and the USEPA. WDNR fish timeframe may not have been analyzed for the were collected as part of regular population same number of PFCs. assessments and fisheries surveys using methods appropriate for each fish species and A brief description of analytical techniques waterbody type (i.e. seining, electroshocking, used by WSLH is given here: 0.5 g of partially gill netting, etc.). Collection locations were thawed homogenized sub-sample and 0.5 mL chosen in order to capture PFC variability in of 18 Mohm·cm water were further 2 Carbon Detection limit chain CAS Years (ng/g) Abbreviation Name Formula length number analyzed NCCA/GL WSLH PFBA Perfluorobutanoate C3F7COOH 3 375-22-4 2006-2012 0.07 0.12 – 3.7 PFPeA Perfluoropentanoate C4F9COOH 4 2706-90-3 2006-2012 0.13 0.12 – 2.5 PFBS Perfluorobutane sulfonate C4F9SO3 4 375-73-5 2006-2012 0.10 0.12 – 5.0 PFHxA Perfluorohexanoate C5F11COOH 5 307-24-4 2006-2012 0.07 0.12 – 2.5 PFHpA Perfluoroheptanoate C6F13COOH 6 375-85-9 2006-2012 0.09 0.12 – 2.5 PFHxS Perfluorohexane sulfonate C6F13SO3 6 355-46-4 2006-2012 0.12 0.50 – 5.0 PFOA Perfluorooctanoate C7F15COOH 7 335-67-1 2006-2012 0.10 0.12 – 2.5 PFHpS* Perfluoro-1-heptanesulfonate C7HF15SO3 7 375-92-8 2009-2012 n/a 0.12 – 0.50 PFNA Perfluorononanoate C8F17COOH 8 375-95-1 2006-2012 0.08 0.12 – 2.5 PFOSA Perfluorooctane sulfonamide C8F17SO2NH2 8 754-91-6 2006-2010 0.08 2.26 – 2.5 PFOS Perfluorooctanesulfonate C8F17SO3 8 1763-23-1 2006-2012 0.13 0.12 – 5.0 PFDA Perfluorodecanoate C9F19COOH 9 335-76-2 2006-2012 0.06 0.12 – 2.5 PFUnA Perfluoroundecanoate C10F21COOH 10 2058-94-8 2006-2012 0.11 0.12 – 2.5 PFDS* Perfluoro-1-decanesulfonate C10F21SO3 10 335-77-3 2009-2012 n/a 0.12 – 0.50 PFDoA Perfluorododecanoate C11F23COOH 11 307-55-1 2006-2012 0.12 0.12 – 2.5 PFTrDA* Perfluoro-n-tridecanoic acid C13HF25O2 13 72629-94-8 2010-2012 n/a 0.12 – 0.50 PFTeDA* Perfluoro-n-tetradecanoic acid C14HF27O2 14 376-06-7 2010-2012 n/a 0.12 – 0.50 * PFC analyzed only by WI State Laboratory of Hygiene homogenized in 50 mL polypropylene tubes final extract was syringe filtered using 13 mm, using a probe mixer. Nine mL of 10 mM 0.2 μm Millex-GN® nylon discs into sodium hydroxide in methanol was used to autosampler vials and capped with aluminum rinse residual homogenate from the mixer crimp caps containing polypropylene septa. probe while being added to the tube.